Analog Unidirectional Memristive and Memcapacitive Device for Neuromorphic Computing

Intelligent and adaptable systems can greatly benefit from the capabilities of tunable electronics. Memristors and memcapacitors have undergone thorough investigation as compelling options for fulfilling the demand of various emerging applications, thanks to their attributes of low power consumption, high density, and rapid operation, enabling the provision of adjustable states. This work focused on the characteristics of analog unidirectional memristive and memcapacitive switching in ITO/ZnO/Yb2O3/Au. The analog resistive memory device exhibited a gradual transformation in resistance and capacitance during repeated voltage sweeps. Simultaneously, the coexistent analog memristive and memcapacitive traits emulated the dynamics of biological synaptic strengthening and weakening, implying a potential application in neuromorphic devices to analyze long-term memory (LTM) or short-term memory (STM). The device's capacity for altering its weight was used for hardware implementation of neural networks, achieving an impressive accuracy rate of 93%.